Computer architecture containing processor and decoupled coprocessor

a technology of coprocessor and processor, applied in the field of computer architecture, can solve the problems of inability to optimize for all tasks, inflexible use, and cpu failures, and achieve the effect of great efficiency

Inactive Publication Date: 2008-06-03
HEWLETT-PACKARD ENTERPRISE DEV LP
View PDF33 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Benefits of technology

[0012]Where such a buffer memory is used, and as the first processor is decoupled from the other system elements, it is desirable for there to be a synchronisation mechanism to synchronise transfer of data between the buffer memory and the memory with execution of instructions by the second processor. Preferably, this is adapted to block execution of instructions by the second processor on data which has not yet been loaded to the buffer memory from the memory, and is adapted to block execution memory instructions for storage of data from the buffer memory to the memory where relevant instructions have not yet been executed by the second processor. Greatest efficiency is achieved when if execution of instructions or memory instructions is blocked by the synchronisation mechanism, other instructions or memory instructions which are not blocked by the synchronisation mechanism may still be carried out.

Problems solved by technology

Such microprocessors are well adapted to handle a wide range of computational tasks, but they are inevitably not optimised for all tasks.
Where tasks are computationally intense (such as media processing) then the CPU will frequently not be able to perform acceptably.
They are however inflexible both in use and in programming (as they are designed for a specific task alone) and are typically slow to produce.
Although use of such coprocessors can considerably improve the efficiency of such computation, the limitations of the microprocessor acting as CPU can still have a very significant effect on overall system performance where such computations are required.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Computer architecture containing processor and decoupled coprocessor
  • Computer architecture containing processor and decoupled coprocessor
  • Computer architecture containing processor and decoupled coprocessor

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028]FIG. 1 shows the basic elements of a system in accordance with a first embodiment of the invention. Essentially, the system comprises a processor 1 and a coprocessor 2, established so that a calculation can be partitioned between the processor 1 and the coprocessor 2 for greatest computational efficiency. The processor 1 may be essentially any general purpose processor (for example, an i960) and the coprocessor 2 essentially any coprocessor capable of handling with significantly greater effectiveness a part of the calculation. In the specific system described here, essentially the whole computation is to be handled by the coprocessor 2, rather than by the processor 1—however, the invention is not limited to this specific arrangement.

[0029]In the system specifically described, coprocessor 2 is a form of reconfigurable FPGA, as will be discussed further below—however, other forms of coprocessor 2, such as, for example, ASICS and DSPs, could be employed instead (with correspondin...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

A computer system comprises a first processor 1 and a second processor 2 for use as a coprocessor to the first processor 1. The system has a main memory 3. The system also has a decoupling element 8 such that instructions are passed to the second processor 2 from the first processor 1 through the decoupling element 8. This has the effects that the second processor 2 consumes instructions derived from the first processor 1 through the decoupling element 8, and that the second processor 2 receives data from and writes data to the memory 3. The processing of instructions by the second processor 2 can thus be decoupled from the operation of the first processor 1.
This is particularly effective for processing of a computationally intensive task (such as a media computation) on an architecture with a general purpose first processor 1, using a second processor 2 adapted for the computationally intensive task. This can effectively be combined with use of a buffer memory 5 adapted to exchange data particularly rapidly with the memory 3 in response to memory instructions, together with a further decoupling element 6 to decouple the buffer memory 5 from the first processor 1.

Description

FIELD OF INVENTION[0001]The invention relates to computer architectures involving a main processor and a coprocessor.DESCRIPTION OF PRIOR ART[0002]Microprocessor-based computer systems are typically based around a general purpose microprocessor as CPU. Such microprocessors are well adapted to handle a wide range of computational tasks, but they are inevitably not optimised for all tasks. Where tasks are computationally intense (such as media processing) then the CPU will frequently not be able to perform acceptably.[0003]One of the possible approaches to this problem is to use coprocessors specifically adapted to handle individual computationally difficult tasks. Such coprocessors are termed ASICs (Application Specific Integrated Circuits). These are built for specific computational tasks, and can thus be optimised for such tasks. They are however inflexible both in use and in programming (as they are designed for a specific task alone) and are typically slow to produce. Improved so...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(United States)
IPC IPC(8): G06F15/00
CPCG06F15/16
Inventor OLGIATI, ANDREAMCCARTHY, DOMINIC PAUL
Owner HEWLETT-PACKARD ENTERPRISE DEV LP
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products